Protein folding is a spontaneous process that is driven by the small difference in Gibbs free energy between the native and unfolded state. Within the folding process, a largely unstructured polypeptide chain adopts what is termed the native conformation or three-dimensional structure of a protein. Aggregation of incompletely folded molecules competes with productive folding, and this constitutes a major problem and affects the folding yields both in vivo and in vitro. In living cells, folding is assisted by helper proteins. Folding helpers are polypeptides that assist the folding and maintain the structural integrity of other proteins. They possess the ability to promote the proper folding of a polypeptide chain by reversibly interacting with their target, thereby preventing detrimental side reactions such as aggregation processes. They do so both in vivo and in vitro, and there is an ever increasing number of applications of these folding helpers in biotechnological problems. Generally, folding helpers are subdivided into folding catalysts and chaperones.
Chaperones are known to reversibly bind to denatured, partially denatured or, put simply, hydrophobic surfaces of polypeptides and thus help to renature proteins or to keep them in solution. Chaperones lower the concentration of aggregation-prone folding intermediates and aggregation-prone folded proteins by reversibly binding and masking hydrophobic surfaces. They thus exert a mere binding function. In contrast, folding catalysts such as disulfide oxidoreductases and peptidyl-prolyl cis/trans isomerases accelerate rate limiting steps in protein folding and thus shorten the lifetime of folding intermediates. Folding catalysts thus lower the concentration of aggregation-prone folding intermediates due to their catalytic function. An important class of folding catalysts is referred to as peptidyl prolyl cis/trans isomerases (PPIases).
Based on sequence similarity, protein topology and binding of immnunosuppressant molecules, prolyl isomerases are distinguished into three distinct families, the cyclophilins, the parvulins and the FK506 binding proteins (hence the acronym FKBPs). FKBPs bind to and are inhibited by FK506, rapamycin and related macrolide derivatives, which have been used as immunosuppressant drugs.
A putative folding helper that belongs to the FKBP family of peptidyl prolyl cis/trans isomerases in E. coli is SlpA, SlpA being the acronym for “SlyD-like protein A” (Hottenrott et al. 1997, JBC 272/25, 15697-15701). Up to now, information on SlpA and its physiological role in E. coli has been scarce. Although a poor prolyl isomerase activity of SlpA has been reported, this protein has hitherto remained rather enigmatic. So far, information on the physico-chemical or possible chaperone properties of SlpA has been lacking, and the function of SlpA in the E. coli cytosol has not even been addressed.
In many diagnostic applications recombinantly produced proteins are used as binding partners, e.g., as antigens in an immunoassay designed for the detection of a specific immunoglobulin analyte. These antigens may be produced as fusion proteins containing one part that makes up the target portion or antigenic polypeptide which is intended to recognize and bind a specific moiety present in the sample or in the assay mixture under study. The other part of the recombinantly produced fusion protein is a polypeptide portion that is fused to the specificity-conferring antigenic part in order to facilitate its cloning, expression, overproduction, folding/refolding and purification, and to increase its solubility, its stability or its reversibility of folding. The synthesis of recombinantly produced fusion proteins is well described in prior art. It is also well-established that it is advantageous to use chaperones as that part of the fusion protein that serves a role as a helping molecule for the expression, folding, purification, solubilization, and the increase in the overall stability of the target polypeptide.
U.S. Pat. No. 6,207,420 discloses a fusion protein system for the expression of proteins, in which the amino acid sequences of the target polypeptide part and the fused peptide part originate from different organisms. Recently it could be shown that FkpA and SlyD are suitable as fusion modules for the production of recombinant proteins. Both chaperones increase the expression rate of their client proteins in a prokaryotic host, support correct refolding and increase the overall solubility of even extremely aggregation-prone proteins such as retroviral transmembrane proteins (Scholz et al. 2005, JMB 345, 1229-1241 and WO 03/000877).
While FkpA and SlyD are particularly useful in helping difficult or aggregation-prone proteins to adopt and maintain their native structure in diagnostic reagents and, more generally speaking, biotechnological applications, there remains the challenge of thermal stability. The native conformation of proteins is stabilized by a carefully balanced network of van-der-Waals contacts, hydrogen bonds, salt bridges and hydrophobic interactions. These contacts are optimized for the microenvironment of the respective protein, and changes in pH, ionic strength or temperature do perturb and shift the equilibrium between folded and unfolded molecules. An increase in temperature is particularly well suited to denature proteins, which often results in aggregation of the fully or partially unfolded molecules. Thermally induced aggregation of proteins with the concomitant loss of function constitutes a major problem of any protein formulation. It is well conceivable that elevated temperatures, or, more generally speaking, thermal stress may occur during inappropriate shipment or storage of protein reagents or formulations.
A chaperone fusion module such as SlyD, for instance, shows an onset of thermally induced unfolding at a temperature around 42° C., a temperature which is easily exceeded, e.g., when the cooling system is defective in a container used for transportation, shipment or storage of a protein formulation. In case the target protein X is highly hydrophobic and fully depends on the chaperoning activity of its fusion partner, the complete fusion polypeptide will aggregate as soon as the SlyD module unfolds and concomitantly loses its solubilizing function. In other words, the stability of SlyD limits the overall stability of a SlyD-X fusion polypeptide when X is a very hydrophobic and aggregation-prone client protein.
Fusion proteins comprising FkpA show a slightly increased stability, probably due to the higher intrinsic thermostability of the dimeric FkpA carrier module. The melting temperature of E. coli SlyD has been determined at around 42° C., whereas FkpA is rather stable up to around 50° C. Yet, for reasons that are outlined in the following section, there remains the urgent need to provide alternative functional chaperone variants with high intrinsic stability.
In a heterogeneous immunoassay of the double antigen sandwich (DAGS) format, for instance, two variants of an antigen are employed on either side of the assay. One of these variants bears a label with a high affinity for the solid phase, the other bears a signaling moiety in order to generate a signal output. Each of these antigen variants may be fused to a helper sequence, i.e., a carrier or fusion module. At least one chaperone (or a functional polypeptide binding domain, i.e. a chaperone domain) is attached or fused to the target polypeptide and facilitates folding, increases stability and solubility and maintains the target polypeptide in a suitable conformation so that the antibody analyte to be determined can specifically recognize and bind the target polypeptide. Preferably, different chaperones are used as fusion partners on either side of an immunological bridge assay, in order to break the inherent symmetry of the assay. An assay format containing different carrier or fusion modules but identical or similar target polypeptides on either side (i.e., on the capture and the detection side) may also be termed an asymmetric DAGS format. By using different fusion modules on each side of a DAGS assay, the risk of immunological cross-reactions due to the carrier modules and, concomitantly, erroneously high signals may be substantially reduced.
Clearly, the overall stability of the assay is limited by the immunological component with the lowest inherent stability. When using FkpA and SlyD as fusion partners in an asymmetric DAGS, SlyD is the fusion partner that limits the overall stability. Thus, there is an obvious need to find other chaperones, which can fully replace SlyD functionally and which are inherently more stable towards thermal stress. Even though a wealth of SlyD homologues from thermophilic or hyperthermophilic organisms have been described, there is a caveat in simply using these proteins as fusion partners: Since they have been evolved and optimized for temperatures far beyond 60° C. they possess an extremely high thermodynamic stability. As a consequence, stable and hyperstable proteins often tend to become rather rigid at ambient temperature, i.e., they lose the flexibility which is a prerequisite for dynamic binding to target molecules. It is widely accepted that the stability of a protein can only be increased at the expense of both its flexibility and function, which often precludes highly stable proteins from applications at ambient temperature. An object of the present invention is therefore to identify thermostable folding helpers from mesophilic organisms. A further object of the present invention is to provide polypeptides suitable for diagnostic and biotechnological applications that possess an increased thermal stability and prolong the shelf life of diagnostic reagents and protein formulations.
A few proteins of E. coli are stable and soluble at temperatures far beyond 49° C. as reported recently by Kwon et al. (BMB reports 2008, 41(2), 108-111). The proteins that were soluble upon exposure to elevated temperature were identified by SDS polyacrylamide gel electrophoresis. The study was carried out with sonicated extracts of E. coli after incubation at various temperatures. Amongst the 17 heat-stable proteins that were identified, 6 proteins turned out to be putative folding helpers (GroEL, GroES, DnaK, FkpA, trigger factor, EF-Ts). It is noteworthy that the experiment was performed with a cell-free lysate of E. coli and that the solubility of the respective protein was taken as the sole criterion for stability.
There is, however, a substantial difference between the solubility and the stability of a protein. It is well known in the art that the solubility of a protein often reaches a minimum at conditions of maximal stability. For instance, the thermodynamic stability of a protein reaches a maximum when the pH of the buffer solution coincides with the pI of the respective protein. Yet, under these very conditions, the protein solubility reaches a minimum. Another popular example is the salting-out of proteins by means of ammonium sulfate or other cosmotropic agents: here also, the solubility of a protein decreases as its stability is increased (ammonium sulfate is a strongly cosmotropic agent, i.e., it stabilizes protein structures).
WO 2007/077008 discloses recombinantly produced chimeric fusion proteins that contain the polypeptide binding segment of a non-human chaperone like, e.g., E. coli SlyD, and N- and C-terminally fused thereto a human FKBP type peptidyl-prolyl-cis/trans isomerase. A similar fusion polypeptide has been disclosed using a chaperone segment of SlpA.
Surprisingly, SlpA, in particular E. coli SlpA, is able to confer thermal stability on other target polypeptides when used as a fusion partner. As reported by Hottenrott et al. (supra) SlpA is an enzyme with a rather poor peptidyl-prolyl cis/trans isomerase activity. Unexpectedly, SlpA exhibits also pronounced chaperone features and, even more surprisingly, SlpA possesses an high intrinsic stability and confers thermal stability on a fused target polypeptide thereby making the target polypeptide less susceptible to heat-induced aggregation. Whereas the closely related SlyD exhibits only a marginal stability with a midpoint of thermal unfolding at around 42° C., SlpA retains its native fold at least up to 50° C. and shows a midpoint of thermal unfolding (defined as the melting temperature) at around 56° C. This is indeed puzzling given the close relationship between SlyD and SlpA (which stands for SlyD-like protein) and given the fact that both are monomeric proteins from a mesophilic organism such as E. coli with a maximum growth temperature of 49° C. Most surprisingly, the mesophilic organism E. coli harbors a putative folding helper such as SlpA that combines outstanding thermostability and chaperone features.